Jan Leyonberg 3ebd79751f [MLIR][ROCDL] Remove patterns for ops supported as intrinsics in the AMDGPU backend (#102971) ...

This patch removes patterns for a few operations which allows mathToLLVM
conversion to convert the operations into LLVM intrinsics instead since
they are supported directly by the AMDGPU backend.

2024-09-04 11:29:10 -04:00

..

AffineToStandard

Reapply "[mlir] NFC: fix dependence of (Tensor|Linalg|MemRef|Complex) dialects on LLVM Dialect and LLVM Core in CMake build (#104832)" (#105703)

2024-08-28 22:34:14 -06:00

AMDGPUToROCDL

[MLIR][AMDGPU] Add support for fp8 ops on gfx12 (#106388)

2024-09-03 17:47:08 +01:00

ArithCommon

[mlir][llvm] Port overflowFlags to a native operation property (RELAND) (#89410)

2024-04-19 09:23:00 -07:00

ArithToAMDGPU

[MLIR][AMDGPU] Introduce fp16 packed arithmetic (#105688)

2024-08-26 12:48:57 -05:00

ArithToArmSME

[mlir][ArmSME] Rename slice move operations to insert/extract_tile_slice (#106755)

2024-09-02 11:12:40 +01:00

ArithToEmitC

[mlir][emitc] Lower arith.divui, remui (#99313)

2024-07-31 10:41:18 +02:00

ArithToLLVM

[MLIR][Arith] Add rounding mode attribute to truncf (#86152)

2024-04-01 11:57:14 +02:00

ArithToSPIRV

[mlir][arith][spirv] Convert arith.truncf rounding mode to SPIR-V (#101547)

2024-08-02 10:30:43 -04:00

ArmNeon2dToIntr

…

ArmSMEToLLVM

[mlir][ArmSME] Rename slice move operations to insert/extract_tile_slice (#106755)

2024-09-02 11:12:40 +01:00

ArmSMEToSCF

[mlir][ArmSME] Rename slice move operations to insert/extract_tile_slice (#106755)

2024-09-02 11:12:40 +01:00

AsyncToLLVM

…

BufferizationToMemRef

[mlir] [bufferize] fix bufferize deallocation error in nest symbol table (#98476)

2024-07-15 12:52:46 +08:00

ComplexToLibm

…

ComplexToLLVM

Reapply "[mlir] NFC: fix dependence of (Tensor|Linalg|MemRef|Complex) dialects on LLVM Dialect and LLVM Core in CMake build (#104832)" (#105703)

2024-08-28 22:34:14 -06:00

ComplexToSPIRV

…

ComplexToStandard

Fix complex abs with nnan/ninf. (#95080)

2024-06-11 13:44:29 +02:00

ControlFlowToLLVM

…

ControlFlowToSCF

Reapply "[mlir] NFC: fix dependence of (Tensor|Linalg|MemRef|Complex) dialects on LLVM Dialect and LLVM Core in CMake build (#104832)" (#105703)

2024-08-28 22:34:14 -06:00

ControlFlowToSPIRV

mlir/LogicalResult: move into llvm (#97309)

2024-07-02 10:42:33 +01:00

ConvertToLLVM

[mlir] Support DialectRegistry extension comparison (#101119)

2024-08-06 01:32:36 +02:00

ConvertToSPIRV

[mlir][spirv] Integrate convert-to-spirv into mlir-vulkan-runner (#106082)

2024-08-27 12:16:54 -04:00

FuncToEmitC

[mlir][EmitC] Fix call ops with zero arguments in func to emitc conversion (#94936)

2024-06-10 13:40:24 +02:00

FuncToLLVM

[mlir][LLVM] Improve lowering of llvm.byval function arguments (#100028)

2024-08-08 19:27:54 -07:00

FuncToSPIRV

mlir/LogicalResult: move into llvm (#97309)

2024-07-02 10:42:33 +01:00

GPUCommon

[MLIR][GPU-LLVM] Add GPU to LLVM-SPV address space mapping (#102621)

2024-08-16 11:18:35 +02:00

GPUToLLVMSPV

[mlir][gpu] Add extra value types for gpu::ShuffleOp (#104605)

2024-08-20 19:50:25 +01:00

GPUToNVVM

[mlir][gpu] Add 'cluster_size' attribute to gpu.subgroup_reduce (#104851)

2024-08-20 13:37:03 -04:00

GPUToROCDL

[mlir][GPU] Improve gpu.module op implementation (#102866)

2024-08-13 09:37:36 +02:00

GPUToSPIRV

[mlir][gpu] Add 'cluster_size' attribute to gpu.subgroup_reduce (#104851)

2024-08-20 13:37:03 -04:00

GPUToVulkan

…

IndexToLLVM

…

IndexToSPIRV

…

LinalgToStandard

[MLIR] Add missing MLIRLinalgTransforms to LinalgToStandard conv (#84545)

2024-03-20 09:26:16 +00:00

LLVMCommon

[mlir][LLVM] Improve lowering of llvm.byval function arguments (#100028)

2024-08-08 19:27:54 -07:00

MathToFuncs

…

MathToLibm

[MLIR][MathDialect] fix fp32 promotion crash when encounters scf.if (#104451)

2024-08-21 10:08:56 +08:00

MathToLLVM

…

MathToROCDL

[MLIR][ROCDL] Remove patterns for ops supported as intrinsics in the AMDGPU backend (#102971)

2024-09-04 11:29:10 -04:00

MathToSPIRV

[mlir][[spirv] Add support for math.log2 and math.log10 to GLSL/OpenCL SPIRV Backends (#104608)

2024-08-19 11:39:52 -04:00

MemRefToEmitC

[mlir][EmitC] Model lvalues as a type in EmitC (#91475)

2024-08-20 11:52:16 +02:00

MemRefToLLVM

mlir: fix incorrect usages of divideCeilSigned (#95680)

2024-06-17 11:43:57 +01:00

MemRefToSPIRV

mlir/LogicalResult: move into llvm (#97309)

2024-07-02 10:42:33 +01:00

NVGPUToNVVM

[mlir][nvgpu] Add nvgpu.rcp OP (#100965)

2024-07-30 09:20:49 +02:00

NVVMToLLVM

mlir/LogicalResult: move into llvm (#97309)

2024-07-02 10:42:33 +01:00

OpenACCToSCF

…

OpenMPToLLVM

[MLIR][OpenMP] Add missing OpenMP to LLVM conversion patterns (#104440)

2024-08-21 16:57:31 +01:00

PDLToPDLInterp

Add llvm::min/max_element and use it in llvm/ and mlir/ directories. (#84678)

2024-03-10 20:00:13 -07:00

ReconcileUnrealizedCasts

[mlir][Transforms][NFC] Move ReconcileUnrealizedCasts implementation (#104671)

2024-08-23 08:46:31 -07:00

SCFToControlFlow

[mlir] fix missing LLVMDialect dependency for MLIRSCFToControlFlow

2024-08-28 23:27:50 -06:00

SCFToEmitC

[mlir][EmitC] Model lvalues as a type in EmitC (#91475)

2024-08-20 11:52:16 +02:00

SCFToGPU

…

SCFToOpenMP

[MLIR][OpenMP] NFC: Sort clauses alphabetically (1/2) (#101193)

2024-07-31 10:40:11 +01:00

SCFToSPIRV

[mlir][Transforms] Dialect Conversion: Simplify block conversion API (#94866)

2024-06-10 21:49:52 +02:00

ShapeToStandard

…

SPIRVCommon

[MLIR][GPU-LLVM] Convert gpu.func to llvm.func (#101664)

2024-08-09 16:09:11 +02:00

SPIRVToLLVM

[MLIR][GPU-LLVM] Convert gpu.func to llvm.func (#101664)

2024-08-09 16:09:11 +02:00

TensorToLinalg

mlir/LogicalResult: move into llvm (#97309)

2024-07-02 10:42:33 +01:00

TensorToSPIRV

[mlir][TensorToSPIRV] Add type check for tensor.extract in TensorToSPIRV (#107110)

2024-09-04 10:21:27 +08:00

TosaToArith

…

TosaToLinalg

[TOSA] Add lowering for tosa.sin and tosa.cos (#99651)

2024-07-19 16:11:23 +01:00

TosaToMLProgram

…

TosaToSCF

…

TosaToTensor

TosaToTensor: Support reshape on tensors of unsigned integer (#91734)

2024-05-28 17:59:23 +02:00

UBToLLVM

…

UBToSPIRV

…

VectorToArmSME

[mlir][ArmSME] Rename slice move operations to insert/extract_tile_slice (#106755)

2024-09-02 11:12:40 +01:00

VectorToGPU

[mlir] Construct SmallVector with ArrayRef (NFC) (#101896)

2024-08-04 11:43:05 -07:00

VectorToLLVM

Reapply "[mlir] NFC: fix dependence of (Tensor|Linalg|MemRef|Complex) dialects on LLVM Dialect and LLVM Core in CMake build (#104832)" (#105703)

2024-08-28 22:34:14 -06:00

VectorToSCF

Reapply "[mlir] NFC: fix dependence of (Tensor|Linalg|MemRef|Complex) dialects on LLVM Dialect and LLVM Core in CMake build (#104832)" (#105703)

2024-08-28 22:34:14 -06:00

VectorToSPIRV

[mlir][vector] Use DenseI64ArrayAttr for shuffle masks (#101163)

2024-07-30 15:00:14 +01:00

CMakeLists.txt

[MLIR][GPU-LLVM] Convert gpu.func to llvm.func (#101664)

2024-08-09 16:09:11 +02:00